An accurate geomorphometric description of the Iranian loess plateau landscape will further enhance our understanding of recent and past geomorphological processes in this strongly dissected landscape. Therefore, four different input datasets for four landform classification methods were used in order to derive the most accurate results in comparison to ground-truth data from a geomorphological field survey. The input datasets in 5 m and 10 m pixel resolution were derived from Pléiades stereo satellite imagery and the “Shuttle Radar Topography Mission” (SRTM), and “Advanced Spaceborne Thermal Emission and Reflection Radiometer” (ASTER GDEM) datasets with a spatial resolution of 30 m were additionally applied. The four classification approaches tested with this data include the stepwise approach after Dikau, the geomorphons, the topographical position index (TPI) and the object based approach. The results show that input datasets with higher spatial resolutions produced overall accuracies of greater than 70% for the TPI and geomorphons and greater than 60% for the other approaches. For the lower resolution datasets, only accuracies of about 40% were derived, 20–30% lower than for data derived from higher spatial resolutions. The results of the topographic position index and the geomorphons approach worked best for all selected input datasets.

Many geoscientific computations are directly influenced by the resolution and accuracy of digital elevation models (DEMs). Therefore, knowledge about the accuracy of DEMs is essential to avoid misleading results. In this study, a comprehensive evaluation of the vertical accuracy of globally available DEMs from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Shuttle Radar Topography Mission (SRTM), Advanced Land Observing Satellite (ALOS) World 3D and TanDEM-X WorldDEM™ was conducted for a large region in Northern Chile. Additionally, several very high-resolution DEM datasets were derived from Satellite Pour l’Observation de la Terre (SPOT) 6/7 and Pléiades stereo satellite imagery for smaller areas. All datasets were evaluated with three reference datasets, namely elevation points from both Ice, Cloud, and land Elevation (ICESat) satellites, as well as very accurate high-resolution elevation data derived by unmanned aerial vehicle (UAV)-based photogrammetry and terrestrial laser scanning (TLS). The accuracy was also evaluated with regard to the existing relief by relating the accuracy results to slope, terrain ruggedness index (TRI) and topographic position index (TPI). For all datasets with global availability, the highest overall accuracies are reached by TanDEM-X WorldDEM™ and the lowest by ASTER Global DEM (GDEM). On the local scale, Pléiades DEMs showed a slightly higher accuracy as SPOT imagery. Generally, accuracy highly depends on topography and the error is rising up to four times for high resolution DEMs and up to eight times for low-resolution DEMs in steeply sloped terrain compared to flat landscapes.

This study evaluates the vertical accuracy of nine freely available digital elevation models (DEMs) with a medium spatial resolution for a typical mid-latitude flat to hilly landscape in the Rur catchment in Germany. All datasets were evaluated with two reference datasets, a 1 m DEM and a highly precise set of elevation control points. The accuracy was evaluated with regard to different slopes, landforms and land use. The results reveal in flat areas an average error of 2-4 m which increases about two to three times in undulated terrain. Areas with dense tree cover show a mean error of 6-10 m. The highest accuracies were achieved by the SRTM DEM, followed by other DEMs that mainly implemented SRTM data. TanDEM-X and Copernicus DEM showed ambivalent results showing a higher ratio of existing errors in the elevation product for vegetation and water areas that prevent a higher overall accuracy of these DEMs.

The resolution and accuracy of digital elevation models (DEMs) have direct influence on further geoscientific computations like landform classifications and hydrologic modelling results. Thus, it is crucial to analyse the accuracy of DEMs to select the most suitable elevation model regarding aim, accuracy and scale of the study. Nowadays several worldwide DEMs are available, as well as DEMs covering regional or local extents. In this study a variety of globally available elevation models were evaluated for an area of about 190,000 km². Data from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) 30 m, Shuttle Radar Topography Mission (SRTM) 30 m and 90 m, Advanced Land Observing Satellite (ALOS) World 3D 30 m and TanDEM-X WorldDEM™ – 12 m and 90 m resolution were obtained. Additionally, several very high resolution DEM data were derived from stereo satellite imagery from SPOT 6/7 and Pléiades for smaller areas of about 100–400 km² for each dataset. All datasets were evaluated with height points of the Geoscience Laser Altimeter System (GLAS) instrument aboard the NASA Ice, Cloud, and land Elevation (ICESat) satellite on a regional scale and with nine very high resolution elevation models from UAV-based photogrammetry on a very large scale. For all datasets the root mean square error (RMSE) and normalized median absolute deviation (NMAD) was calculated. Furthermore, the association of errors to specific terrain was conducted by assigning these errors to landforms from the topographic position index (TPI), topographic roughness index (TRI) and slope. For all datasets with a global availability the results show the highest overall accuracies for the TanDEM-X 12 m (RMSE: 2.3 m, NMAD: 0.8 m). The lowest accuracies were detected for the 30 m ASTER GDEM v3 (RMSE: 8.9 m, NMAD: 7.1 m). Depending on the landscape the accuracies are higher for all DEMs in flat landscapes and the errors rise significantly in rougher terrain. Local scale DEMs derived from stereo satellite imagery show a varying overall accuracy, mainly depending on the topography covered by the scene.

Kenya’s road network significantly influences environmental and socio-economic dynamics. High-quality road data is essential for analyzing its impact on various factors, including land-use, biodiversity, migration, livelihoods, and economy. Like many countries, Kenya faces challenges in the availability of accurate and detailed digital historical road datasets. To address this, we used deep learning techniques to extract Kenya’s road network from 533 historical topographic maps (1:50,000 and 1:100,000 scale) covering the 1950s–1980s. This involved digitizing, georeferencing, and classifying of 20 different road symbols on all maps, then converting and merging them into a seamless dataset. The statistical evaluation was conducted against manually created roads from seven representative map sheets by calculating precision, recall, and F1 score. Our study provides a detailed historical road dataset for Kenya containing over 56,000 km of historical roads. The statistical validation showed an average F1 score of 0.84, indicating a high classification performance. The methodology offers an applicable approach for national-level historic road network mapping, also transferable to other regions, map types or features.